A Game Theoretic Approach to Safeguards Selection and Optimization

The anticipated global expansion of nuclear power promises to place unprecedented demands on the nonproliferation regime, making the optimization of safeguarding resources critical to international security. To that end, a novel safeguards modeling approach is being developed to determine inspector resource allocation strategies that optimize detection probability. The model uses a game theoretic component to compute the resource allocation strategy that optimizes detection probability. This component is coupled to a discrete event simulator, which generates detection probabilities for individual diversion-interdiction strategy pairs and feeds them into the game theory model. The inspector safeguarding strategy is subject to budget constraints, and a background detection probability is introduced to serve as a surrogate for all safeguarding activities not explicitly modeled. To verify the feasibility of this model, a proof-of concept has been developed using a simple simulation model that calculates detection probabilities for an insider threat scenario, and couples the results to a two-person zero-sum game. A sensitivity analysis was conducted to assess the effect of changes in budget and background detection probability on defender and attacker strategies, as well as overall detection probability.